Publications of S.B. Serpico

Result of the query in the list of publications :

• 5 Articles
• 9 Conference articles
• 5 Technical and Research Reports

new search | get the bibtex

navigation : 1 - 2 display : 10 per page | 20 per page | 50 per page

9 Conference articles

6 - High resolution SAR-image classification by Markov random fields and finite mixtures. G. Moser and V. Krylov and S.B. Serpico and J. Zerubia. In Proc. of SPIE (IS&T/SPIE Electronic Imaging 2010), Vol. 7533, pages 753308, San Jose, USA, January 2010. Keywords : SAR image classification, Dictionary, amplitude probability density, Stochastic EM (SEM), Markov random field, copula. Copyright : SPIE @INPROCEEDINGS{moserSPIE2010a, author = {Moser, G. and Krylov, V. and Serpico, S.B. and Zerubia, J.}, title = {High resolution SAR-image classification by Markov random fields and finite mixtures}, year = {2010}, month = {January}, booktitle = {Proc. of SPIE (IS&T/SPIE Electronic Imaging 2010)}, volume = {7533}, pages = {753308}, address = {San Jose, USA}, url = {http://proceedings.spiedigitallibrary.org/proceeding.aspx?articleid=776565}, pdf = {http://hal.archives-ouvertes.fr/inria-00442348/en/}, keyword = {SAR image classification, Dictionary, amplitude probability density, Stochastic EM (SEM), Markov random field, copula} } Abstract : In this paper we develop a novel classification approach for high and very high resolution polarimetric synthetic aperture radar (SAR) amplitude images. This approach combines the Markov random field model to Bayesian image classification and a finite mixture technique for probability density function estimation. The finite mixture modeling is done via a recently proposed dictionary-based stochastic expectation maximization approach for SAR amplitude probability density function estimation. For modeling the joint distribution from marginals corresponding to single polarimetric channels we employ copulas. The accuracy of the developed semiautomatic supervised algorithm is validated in the application of wet soil classification on several high resolution SAR images acquired by TerraSAR-X and COSMO-SkyMed.

7 - Dictionary-based probability density function estimation for high-resolution SAR data. V. Krylov and G. Moser and S.B. Serpico and J. Zerubia. In Proc. of SPIE (IS&T/SPIE Electronic Imaging 2009), Vol. 7246, pages 72460S, San Jose, USA, January 2009. Keywords : SAR image, Probability density function, parametric estimation, finite mixture models, Stochastic EM (SEM). Copyright : SPIE @INPROCEEDINGS{KrylovSPIE09, author = {Krylov, V. and Moser, G. and Serpico, S.B. and Zerubia, J.}, title = {Dictionary-based probability density function estimation for high-resolution SAR data}, year = {2009}, month = {January}, booktitle = {Proc. of SPIE (IS&T/SPIE Electronic Imaging 2009)}, volume = {7246}, pages = {72460S}, address = {San Jose, USA}, url = {http://proceedings.spiedigitallibrary.org/proceeding.aspx?articleid=812524}, pdf = {http://hal.archives-ouvertes.fr/inria-00361384/en/}, keyword = {SAR image, Probability density function, parametric estimation, finite mixture models, Stochastic EM (SEM)} } Abstract : In the context of remotely sensed data analysis, a crucial problem is represented by the need to develop accurate models for the statistics of pixel intensities. In this work, we develop a parametric finite mixture model for the statistics of pixel intensities in high resolution synthetic aperture radar (SAR) images. This method is an extension of previously existing method for lower resolution images. The method integrates the stochastic expectation maximization (SEM) scheme and the method of log-cumulants (MoLC) with an automatic technique to select, for each mixture component, an optimal parametric model taken from a predefined dictionary of parametric probability density functions (pdf). The proposed dictionary consists of eight state-of-the-art SAR- specific pdfs: Nakagami, log-normal, generalized Gaussian Rayleigh, Heavy-tailed Rayleigh, Weibull, K-root, Fisher and generalized Gamma. The designed scheme is endowed with the novel initialization procedure and the algorithm to automatically estimate the optimal number of mixture components. The experimental results with a set of several high resolution COSMO-SkyMed images demonstrate the high accuracy of the designed algorithm, both from the viewpoint of a visual comparison of the histograms, and from the viewpoint of quantitive accuracy measures such as correlation coefficient (above 99,5%). The method proves to be effective on all the considered images, remaining accurate for multimodal and highly heterogeneous scenes.

8 - SAR amplitude probability density function estimation based on a generalized Gaussian scattering model. G. Moser and J. Zerubia and S.B. Serpico. In Proc. SPIE Symposium on Remote Sensing, Maspalomas, Gran Canaria, Spain, September 2004. @INPROCEEDINGS{moser2004a, author = {Moser, G. and Zerubia, J. and Serpico, S.B.}, title = {SAR amplitude probability density function estimation based on a generalized Gaussian scattering model}, year = {2004}, month = {September}, booktitle = {Proc. SPIE Symposium on Remote Sensing}, address = {Maspalomas, Gran Canaria, Spain}, url = {http://dx.doi.org/10.1117/12.567853}, keyword = {} }

9 - Finite mixture models and stochastic EM for SAR amplitude probability density function estimation based on a dictionary of parametric families. G. Moser and J. Zerubia and S.B. Serpico. In Proc. IEEE International Geoscience and Remote Sensing Symposium (IGARSS), Anchorage , USA, September 2004. @INPROCEEDINGS{moser2004b, author = {Moser, G. and Zerubia, J. and Serpico, S.B.}, title = {Finite mixture models and stochastic EM for SAR amplitude probability density function estimation based on a dictionary of parametric families}, year = {2004}, month = {September}, booktitle = {Proc. IEEE International Geoscience and Remote Sensing Symposium (IGARSS)}, address = {Anchorage , USA}, url = {http://ieeexplore.ieee.org/xpl/freeabs_all.jsp?arnumber=1368708}, keyword = {} }

top of the page

5 Technical and Research Reports

1 - Classification of very high resolution SAR images of urban areas. A. Voisin and V. Krylov and G. Moser and S.B. Serpico and J. Zerubia. Rapport de recherche 7758, INRIA, October 2011. Keywords : Synthetic Aperture Radar (SAR) image, Supervised classification, Bayesian, finite mixture models, hierarchical Markov random fields, wavelet. @TECHREPORT{RR-7758, author = {Voisin, A. and Krylov, V. and Moser, G. and Serpico, S.B. and Zerubia, J.}, title = {Classification of very high resolution SAR images of urban areas}, year = {2011}, month = {October}, institution = {INRIA}, type = {Rapport de recherche}, number = {7758}, url = {http://hal.inria.fr/docs/00/63/10/38/PDF/RR-7758.pdf}, keyword = {Synthetic Aperture Radar (SAR) image, Supervised classification, Bayesian, finite mixture models, hierarchical Markov random fields, wavelet} } Résumé : Dans le cadre d’une approche face aux risques environnementaux, nous proposons une nouvelle méthode de classification bayésienne supervisée. Celle-ci combine une modélisation statistique des images avec une prise en compte contextuelle via des champs de Markov hiérarchiques. Ce rapport de recherche vise à détailler plus amplement cette modélisation contextuelle, à savoir expliciter le modèle mathématique sur quad-arbre et l’obtention des observations par décomposition en ondelettes de l’image originale. Il met également en exergue certaines modifications apportées en vue d’améliorer la classification finale. Abstract : In the framework of the assessment of environmental risks, we propose herein a new supervised Bayesian classification method. It combines statistical image modeling with a contextual approach via hierarchical Markov random fields. This research report aims to further focus on this kind of contextual classification approach by detailing both the quad-tree mathematical model and the statistics of the observations, obtained by wavelet transform. We therefore introduce modifications to a classical Markovian single-scale algorithm that lead to more accurate classification results.

2 - On the Method of Logarithmic Cumulants for Parametric Probability Density Function Estimation. V. Krylov and G. Moser and S.B. Serpico and J. Zerubia. Research Report 7666, INRIA, July 2011. Keywords : Probability density function, Parameter estimation, generalized gamma distribution, K-distribution, Synthetic Aperture Radar (SAR), Classification. Copyright : INRIA/ARIANA @TECHREPORT{RR-7666, author = {Krylov, V. and Moser, G. and Serpico, S.B. and Zerubia, J.}, title = {On the Method of Logarithmic Cumulants for Parametric Probability Density Function Estimation}, year = {2011}, month = {July}, institution = {INRIA}, type = {Research Report}, number = {7666}, url = {http://hal.archives-ouvertes.fr/inria-00605274/en/}, keyword = {Probability density function, Parameter estimation, generalized gamma distribution, K-distribution, Synthetic Aperture Radar (SAR), Classification} } Résumé : L'estimation de paramètres de fonctions de densité de probabilité est une étape majeure dans le domaine du traitement statistique du signal et des images. Dans ce rapport, nous étudions les propriétés et les limites de l'estimation de paramètres par la méthode des cumulants logarithmiques (MoLC), qui est une alternative à la fois au maximum de vraisemblance (MV) classique et à la méthode des moments. Nous dérivons la condition générale suffisante de consistance forte de l'estimation par la méthode MoLC, qui représente une propriété asymptotique importante de tout estimateur statistique. Grâce à cela, nous démontrons la consistance forte de l'estimation par la méthode MoLC pour une sélection de familles de distributions particulièrement adaptées (mais non restreintes) au traitement d'images acquises par radar à synthèse d'ouverture (RSO). Nous dérivons ensuite les conditions analytiques d'applicabilité de la méthode MoLC à des échantillons générés qui suivent les lois des différentes familles de distribution de notre sélection. Enfin, nous testons la méthode MoLC sur des données synthétiques et réelles afin de comparer les différentes propriétés inhérentes aux différents types d'images, l'applicabilité de la méthode et les effets d'un nombre restreint d'échantillons. Nous avons, en particulier, considéré les distributions gamma généralisée et K. Comme exemple d'application, nous avons réalisé des classifications supervisées d'images médicales à ultrason ainsi que d'images de télédétection acquises par des capteurs RSO. Les résultats obtenus montrent que la méthode MoLC est une bonne alternative à la méthode des moments, bien qu'elle contienne certaines limitations. Elle est particulièrement utile lorsqu'une approche directe par MV n'est pas possible. Abstract : Parameter estimation of probability density functions is one of the major steps in the mainframe of statistical image and signal processing. In this report we explore the properties and limitations of the recently proposed method of logarithmic cumulants (MoLC) parameter estimation approach which is an alternative to the classical maximum likelihood (ML) and method of moments (MoM) approaches. We derive the general sufficient condition of strong consistency of MoLC estimates which represents an important asymptotic property of any statistical estimator. With its help we demonstrate the strong consistency of MoLC estimates for a selection of widely used distribution families originating (but not restricted to) synthetic aperture radar (SAR) image processing. We then derive the analytical conditions of applicability of MoLC to samples generated from several distribution families in our selection. Finally, we conduct various synthetic and real data experiments to assess the comparative properties, applicability and small sample performance of MoLC notably for the generalized gamma and K family of distributions. Supervised image classification experiments are considered for medical ultrasound and remote sensing SAR imagery. The obtained results suggest MoLC to be a feasible yet not universally applicable alternative to MoM that can be considered when the direct ML approach turns out to be unfeasible.

3 - Modeling the statistics of high resolution SAR images. V. Krylov and G. Moser and S.B. Serpico and J. Zerubia. Research Report 6722, INRIA, November 2008. Keywords : Synthetic Aperture Radar (SAR) image, Probability density function, parametric estimation, finite mixture models, Stochastic EM (SEM). Copyright : INRIA/ARIANA, 2008 @TECHREPORT{krylovDSEM08, author = {Krylov, V. and Moser, G. and Serpico, S.B. and Zerubia, J.}, title = {Modeling the statistics of high resolution SAR images}, year = {2008}, month = {November}, institution = {INRIA}, type = {Research Report}, number = {6722}, url = {http://hal.archives-ouvertes.fr/inria-00342681/en/}, pdf = {http://hal.archives-ouvertes.fr/docs/00/35/76/27/PDF/RR-6722.pdf}, keyword = {Synthetic Aperture Radar (SAR) image, Probability density function, parametric estimation, finite mixture models, Stochastic EM (SEM)} } Abstract : In the context of remotely sensed data analysis, a crucial problem is represented by the need to develop accurate models for the statistics of pixel intensities. In this work, we develop a parametric finite mixture model for modelling the statistics of intensities in high resolution Synthetic Aperture Radar (SAR) images. Along with the models we design an efficient parameter estimation scheme by integrating the Stochastic Expectation Maximization scheme and the Method of log-cumulants with an automatic technique to select, for each mixture component, an optimal parametric model taken from a predefined dictionary of parametric probability density functions (pdf). In particular, the proposed dictionary consists of eight most efficient state-of-the-art SAR-specific pdfs: Nakagami, log-normal, generalized Gaussian Rayleigh, Heavy-tailed Rayleigh, Weibull, K-root, Fisher and generalized Gamma. The experiment results with a set of several real SAR (COSMO-SkyMed) images demonstrate the high accuracy of the designed algorithm, both from the viewpoint of a visual comparison of the histograms, and from the viewpoint of quantitive measures such as correlation coefficient (always above 99,5%) . We stress, in particular, that the method proves to be effective on all the considered images, remaining accurate for multimodal and highly heterogeneous images.

4 - SAR Amplitude Probability Density Function Estimation based on a Generalized Gaussian Scattering Model. G. Moser and J. Zerubia and S.B. Serpico. Research Report 5153, INRIA, France, March 2004. Keywords : Synthetic Aperture Radar (SAR), Generalised Gaussians. @TECHREPORT{5153, author = {Moser, G. and Zerubia, J. and Serpico, S.B.}, title = {SAR Amplitude Probability Density Function Estimation based on a Generalized Gaussian Scattering Model}, year = {2004}, month = {March}, institution = {INRIA}, type = {Research Report}, number = {5153}, address = {France}, url = {https://hal.inria.fr/inria-00071430}, pdf = {https://hal.inria.fr/file/index/docid/71430/filename/RR-5153.pdf}, ps = {https://hal.inria.fr/docs/00/07/14/30/PS/RR-5153.ps}, keyword = {Synthetic Aperture Radar (SAR), Generalised Gaussians} } Résumé : En télédetection, un problème important est celui de développer des modèles précis pour representer les statistiques des intensités des pixels. En ce qui concerne les données du type Radar à Synthèse d'Ouverture (RSO), cette modélisation constitue un point capital pour la classification ou le débruitage d'une image, par exemple. Dans ce rapport de recherche, une nouvelle méthode d'estimation paramétrique pour les amplitudes d'images RSO est proposée. Elle tient compte de la nature physique des phénomènes de diffusion qui générent une image RSO en adoptant une modèle de gaussiennes generalisées pour les phénomènes de rétrodiffusion. Une expression, sous forme explicite, de la densité de probabilité de l'amplitude est obtenue et un algorithme spécifique d'estimation des paramètres est proposé afin de pouvoir utiliser le modèle proposé. Une mèthode récente fondée sur les «logs-cumulants» est appliquée, dérivant de l'utilisation d'une transformée de Mellin (à la place de la transformée de Fourier usuelle) dans le calcul des fonctions caractéristiques et de la généralisation des concepts de moment et de cumulant correspondante. Les estimées obtenues par la mèthode des log-cumulants pour le modèle d'amplitude fondé sur des gaussiennes généralisées se révelent être calculables numériquement et également consistantes. Dans ce rapport de recherche, l'approche paramètrique proposée est validée sur diverses images radar RSO (ERS, XSAR, ESAR et des radar aéroportés). Les résultats expérimentaux montrent que la mèthode proposée modèlise mieux la densité de probabilité de l'amplitude que beaucoup de modèles paramétriques proposés précédemment pour les phénomènes de rétrodiffusion. Abstract : In the context of remotely sensed data analysis, an important problem is the development of accurate models for the statistics of the pixel intensities. Focusing on Synthetic Aperture Radar (SAR) data, this modelling process turns out to be a crucial task, for instance, for classification or for denoising purposes. In the present report, an innovative parametric estimation methodology for SAR amplitude data is proposed, which takes into account the physical nature of the scattering phenomena generating a SAR image by adopting a generalized Gaussian (GG) model for the backscattering phenomena. A closed form expression for the corresponding amplitude probability density function (PDF) is derived and a specific parameter estimation algorithm is developed in order to deal with the proposed model. Specifically, the recently proposed «method-of-log-cumulants» (MoLC) is applied, which stems from the adoption of the Mellin transform (instead of the usual Fourier transform) in the computation of characteristic functions, and from the corresponding generalization of the concepts of moment and of cumulant. For the developed GG-based amplitude model, the resulting MoLC estimates turn out to be numerically feasible and are also proved to be consistent. The proposed parametric approach is validated using several real ERS-1, XSAR, ESAR and airborne SAR images and the experimental results prove that the method models the amplitude probability density function better than several previously proposed parametric models for the backscattering phenomena.

5 - Dictionary-based Stochastic Expectation-Maximization for SAR amplitude probability density function estimation. G. Moser and J. Zerubia and S.B. Serpico. Research Report 5154, INRIA, France, March 2004. Keywords : Synthetic Aperture Radar (SAR), Stochastic EM (SEM), Finite mixing model. @TECHREPORT{5154, author = {Moser, G. and Zerubia, J. and Serpico, S.B.}, title = {Dictionary-based Stochastic Expectation-Maximization for SAR amplitude probability density function estimation}, year = {2004}, month = {March}, institution = {INRIA}, type = {Research Report}, number = {5154}, address = {France}, url = {https://hal.inria.fr/inria-00071429}, pdf = {https://hal.inria.fr/file/index/docid/71429/filename/RR-5154.pdf}, ps = {https://hal.inria.fr/docs/00/07/14/29/PS/RR-5154.ps}, keyword = {Synthetic Aperture Radar (SAR), Stochastic EM (SEM), Finite mixing model} } Résumé : En télédetection, un problème vital est le besoin de développer des modèles précis pour représenter les statistiques des intensités des images. Dans ce rapport de recherche, nous traitons le problème de l'estimation de la densité de probabilité de l'amplitude d'une image de type Radar à Synthèse d'Ouverture (RSO). Plusieurs modèles théoriques ou heuristiques, ultilisés pour représenter l'amplitude d'un signal du type RSO, ont été proposés dans la littérature et ce sont révelés être efficaces pour différentes types de classes dans le contexte des cartes d'occupation des sols, rendant ainsi difficile le choix d'une seule densité de probabilité paramétrique. Dans ce rapport de recherche, un algorithme d'estimation innovant est proposé, se fondant sur un modèle de mélange fini pour la densité de probabilité de l'amplitude, les diverses composantes du mélange appartenant à un dictionnaire specifique. La mèthode proposée dans ce rapport intégre, de fa on automatique, les procédures de sélection d'un modèle optimal pour chaque composante, d'estimation de paramètres et d'optimisation du nombre de composantes, en combinant un algorithme EM stochastique et la méthode des logs-cumulants pour l'estimation de la densité de probabilité paramètrique. Des resultats expérimentaux sur plusieurs images RSO réelles sont présentés, montrant ainsi que la mèthode proposée est suffisamment précise pour modéliser les statistiques du signal d'amplitude radar de type RSO. Abstract : In the context of remotely sensed data analysis, a crucial problem is represented by the need to develop accurate models for the statistics of the pixel intensities. In the current research report, we address the problem of parametric probability density function (PDF) estimation in the context of Synthetic Aperture Radar (SAR) amplitude data analysis. Specifically, several theoretical and heuristic models for the PDFs of SAR data have been proposed in the literature, and have been proved to be effective for different land-cover typologies, thus making the choice of a single optimal SAR parametric PDF a hard task. In thia report, an innovative estimation algorithm is proposed, which addresses this problem by adopting a finite mixture model (FMM) for the amplitude PDF, with mixture components belonging to a given dictionary of SAR-specific PDFs. The proposed method automatically integrates the procedures of selection of the optimal model for each component, of parameter estimation, and of optimization of the number of components, by combining the Stochastic Expectation Maximization (SEM) iterative methodology and the recently proposed «method-of-log-cumulants» (MoLC) for parametric PDF estimation for non-negative random variables. Experimental results on several real SAR images are presented, showing the proposed method is accurately modelling the statistics of SAR amplitude data.

top of the page

navigation : 1 - 2 display : 10 per page | 20 per page | 50 per page

These pages were generated by